#gemrb@irc.freenode.net logs for 8 Jun 2009 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[01:15:27] <CIA-2> gemrb: 03mattinm * r6357 10/gemrb/trunk/gemrb/plugins/ (Core/Game.h FXOpcodes/FXOpc.cpp):
[01:15:27] <CIA-2> gemrb: fixed regeneration timings (tested for ring of gaxx and regen; they even stack correctly)
[01:15:27] <CIA-2> gemrb: added a few more ROUND based defines, such as ROUND_SECONDS = 6 to make calcs clearer
[01:15:27] <CIA-2> gemrb: need to double-check that RDP_POINTS works on param1 hp PER SECOND
[01:19:51] <mattinm> i'm guessing just from playing the game (i'll test later) that RPD_POINTS determines how much hp in a round
[01:20:17] <mattinm> therefore, a troll with param1 of 3 would get hp every 3/(ROUND_SECONDS) = 3/6 = 2 seconds
[01:27:22] <nickdaly> The BG2-SOA installer appears to now work! Yay!
[01:54:44] <mattinm> http://www.tgeweb.com/ironworks/baldursgate2/bg2faq.shtml
[02:32:45] <-- nickdaly has left IRC ("Night all!")
[03:42:59] <mattinm> that's a nice little read on some of the working; not much we don't already know, but it does confirm why there is no MXSPLDRU in SoA
[04:13:16] --> pupnik810 has joined #gemrb
[04:59:16] <CIA-2> gemrb: 03mattinm * r6358 10/gemrb/trunk/gemrb/plugins/Core/ (Actor.cpp Actor.h Game.cpp):
[04:59:16] <CIA-2> gemrb: individual combat rounds -- still need CombatCounter properly implemented
[04:59:16] <CIA-2> gemrb: added roundTime and lastInit to Actor class; renamed lasttime to lastattack
[04:59:16] <CIA-2> gemrb: removed StartTurn check in UpdateScripts because the Actors check themselves now
[05:01:12] * pupnik810 wags tail and looks like a dog that has just been shown a card trick
[05:02:28] <mattinm> lol
[05:02:45] <mattinm> combat is sometimes frustrating to work on :/
[05:02:59] <pupnik810> i believe that
[05:03:24] <mattinm> but, it's coming together, so i'm happy :)
[05:04:03] <pupnik810> i am pretty psyched
[05:04:51] <mattinm> hmmm, i wonder how far you can make it in the game now... i think i'll try the dungeon right quick
[05:31:43] <CIA-2> gemrb: 03mattinm * r6359 10/gemrb/trunk/gemrb/plugins/Core/Actor.cpp:
[05:31:43] <CIA-2> gemrb: implemented basic random initiative (d10) that gets a -1 bonus per 8 levels
[05:31:43] <CIA-2> gemrb: not "perfect" but creates randomization while also not hurting higher
[05:31:43] <CIA-2> gemrb: characters as much, which is it's intention
[05:31:43] <CIA-2> gemrb: TODO: implement some sort of stat check?
[05:32:35] <mattinm> hehe, i just had a ranger/cleric in my party go on a rampage :D
[05:48:03] <mattinm> okay, time to get some decent sleep tonight
[06:10:48] <pupnik810> cheers
[06:58:44] <-- pupnik810 has left IRC (Read error: 104 (Connection reset by peer))
[08:27:11] --> lynxlynxlynx has joined #gemrb
[08:27:11] --- ChanServ gives channel operator status to lynxlynxlynx
[08:28:13] <-- |Cable| has left IRC ("Leaving")
[08:33:29] <lynxlynxlynx> cool, yet more fixes
[08:56:01] --> danamin has joined #GemRB
[08:56:40] <-- danamin has left IRC (Client Quit)
[09:03:19] <CIA-2> gemrb: 03lynxlupodian * r6360 10/gemrb-launcher/ (5 files): gemrb-launcher 0.8.5 (Nick Daly)
[09:10:05] <fuzzie> moo.
[09:11:55] <fuzzie> oh dear, another round define :)
[09:29:41] <fuzzie> lynxlynxlynx: did you get around to testing the floor?
[09:32:51] <lynxlynxlynx> 7 2009] [23:45:03] <lynxlynxlynx> ch1 is pretty broken for me :/
[09:32:54] <lynxlynxlynx> :45:14] <lynxlynxlynx> the area overlay is consistent now though :)
[09:32:56] <lynxlynxlynx> brb
[09:37:16] <fuzzie> mattinm's patches added a "FIXME: individual combat counter" comment, but i don't think CombatCounter is per-actor
[09:40:34] <fuzzie> and I guess Avenger will review fx_set_regenerating_state. all seems pretty good although that random initiative doesn't look like it'd work too well.
[09:55:44] <lynxlynxlynx> my test was yesterday, before these two changes
[09:56:44] <fuzzie> sorry, those lines are just from reading the overnight patches
[09:57:10] <fuzzie> I haven't tried the challenge since I made it work, but I wouldn't be surprised to find all of these round-timing patches breaking it.
[09:57:17] <lynxlynxlynx> the bork was manifested in two ways: corpses twitching (as seen elsewhere) and an endless influx of globlins and shapechangers; i didn't get to the other enemies
[09:58:14] <lynxlynxlynx> timers are borked too, yes
[09:58:52] <lynxlynxlynx> illasera takes much longer to appear, but maybe that's correct now (iirc it was coded as 30s)
[09:59:09] <lynxlynxlynx> while aataqah stops the fight practically as soon as it starts
[09:59:23] <fuzzie> It's simple enough to compare, I guess, if I can find some time toda.
[10:29:45] <lynxlynxlynx> combat looks pretty good now
[10:30:07] <fuzzie> yes :)
[10:53:19] <-- barra_away has left IRC ("Verlassend")
[11:05:11] --> pupnik has joined #gemrb
[11:10:37] <pupnik> moorning!
[11:10:46] * pupnik makes coffee and rolls
[11:11:21] * pupnik cleans up the cola bottles and pizza boxes
[11:47:36] <fuzzie> morning
[11:51:14] <CIA-2> gemrb: 03lynxlupodian * r6361 10/gemrb/trunk/gemrb/ (8 files in 8 dirs): added savegame succes feedback
[11:51:37] <lynxlynxlynx> one off the list :)
[11:52:04] <fuzzie> can you fix the loading screen to be accurate now? :-)
[11:52:23] <lynxlynxlynx> accurate?
[11:52:24] <fuzzie> but, yay, thanks
[11:52:39] <fuzzie> gemrb's loading screen gets to 100% when it's only about 50% done
[11:53:01] <fuzzie> i think it doesn't take most of the files into account
[12:00:18] <fuzzie> yes, it just opens the files, unpacks the game and does LoadProgress(100)
[12:00:26] <fuzzie> i guess that is a difficult one
[12:01:27] <pupnik> need a list of files, and update load incrementally?
[12:03:18] <fuzzie> Well, loading a new area should give a progressbar too, and it doesn't *look* like we handle that at all right now, so it's maybe a question of designing a more sensible progressbar system :)
[12:05:55] --> barra_library has joined #gemrb
[12:06:53] <lynxlynxlynx> ieDword fentcount, tentcount <-- i guess this could be used
[12:08:01] <fuzzie> i don't think the bif lookups are the really expensive bit, it's the area decompression
[12:08:07] <fuzzie> but i don't have gemrb right now to look
[12:09:23] <lynxlynxlynx> it wouldn't be able to communicate back anyway
[12:10:11] <fuzzie> I mean, the progress bar in the original game isn't *smooth*, but it doesn't get to 100% and then stall for ages while the actual area load happens.
[12:10:46] <fuzzie> I seem to remember the original game progress being pretty jumpy, really, obviously it was doing whole file loads/decompressions between changes.
[12:13:06] <lynxlynxlynx> yeah, it wasn't smooth at all
[12:14:35] <pupnik> then just set bar to 50 instead
[12:14:51] <pupnik> when it finishes, it disappears
[12:15:05] <fuzzie> or we could fix it properly :)
[12:15:11] <fuzzie> i'll look at it later
[12:18:33] <lynxlynxlynx> i guess ChangeMap could increase it
[12:20:10] <fuzzie> well, it doesn't seem important enough to add hacks for
[12:20:40] <fuzzie> it was mostly just an idle comment from one of the things on my todo list :)
[12:20:50] <-- pupnik has left IRC (kornbluth.freenode.net irc.freenode.net)
[12:20:53] --> pupnik has joined #GemRb
[12:25:57] <fuzzie> 'splitting potions not working' is something I don't know where to start looking for
[12:30:18] <wjp> there are DecreaseStackAmount, IncreaseStackAmount, DragItemAount, OpenItemAmountWindow in GUIINV
[12:30:51] <wjp> the actual work seems to be delegated to GemRB.DragItem
[12:33:19] <-- pupnik has left IRC (kornbluth.freenode.net irc.freenode.net)
[12:33:41] --> pupnik has joined #GemRb
[12:46:51] <-- pupnik has left IRC (kornbluth.freenode.net irc.freenode.net)
[12:47:02] --> pupnik has joined #GemRb
[13:08:37] <-- pupnik has left IRC (kornbluth.freenode.net irc.freenode.net)
[13:08:44] --> pupnik has joined #GemRb
[13:10:06] <-- pupnik has left IRC (kornbluth.freenode.net irc.freenode.net)
[13:25:42] <fuzzie> wjp: thanks
[13:37:46] <-- jdm has left IRC ("http://www.mibbit.com ajax IRC Client")
[14:02:59] <CIA-2> gemrb: 03fuzzie * r6362 10/gemrb/trunk/gemrb/GUIScripts/bg2/GUIINV.py: don't force-drop items in OpenItemAmountWindow if we're actually closing the window
[14:03:01] <fuzzie> i don't understand why we're calling OpenItemAmountWindow in that situation anyway, maybe someone wants to review :)
[14:22:57] <fuzzie> it would be nice if a commit like "some more default/cancel buttons" didn't add mysterious code :/
[14:23:35] <lynxlynxlynx> i guess that's from avenger, huh
[14:23:38] <fuzzie> yes :)
[14:25:24] <fuzzie> i guess this is an attempt to sync guiscripts, but it didn't work
[14:37:16] <fuzzie> http://fuzzie.org/nfs/gemrb/pst_fix_splititems.txt
[14:37:46] <fuzzie> ^- this fixes splitting items in PS:T, but i don't understand why (i copied the code from elsewhere in GUIINV and from bg2's GUIINV), so i don't know if it's correct
[14:39:10] <fuzzie> i guess it makes it no worse, so should probably just commit
[14:41:28] <CIA-2> gemrb: 03fuzzie * r6363 10/gemrb/trunk/gemrb/GUIScripts/pst/GUIINV.py: randomly copy some pst/bg2 GUIINV code to fix item splitting in pst
[14:47:21] <lynxlynxlynx> heh
[14:52:40] <fuzzie> the guiscripts are really still just black magic to me
[14:54:08] --> pupnik_ has joined #gemrb
[15:05:24] <-- pupnik_ has left IRC ("Leaving")
[15:05:37] --> pupnik has joined #gemrb
[15:06:14] <-- pupnik has left IRC (Client Quit)
[15:07:09] --> pupnik has joined #gemrb
[15:31:11] <fuzzie> the little manual included with TotSC has some kind of hilarious patch notes
[15:31:51] <fuzzie> "a thief's ability to hide in shadows will now be affected by the presence (or absence) of shadows", "a thief's ability to backstab an opponent now requires the thief to be behind the target", etc
[15:32:24] <fuzzie> funny to remember how primitive bg1 was in a lot of ways
[15:36:02] <pupnik> mhm
[15:36:09] <pupnik> lots of rules
[15:36:15] <lynxlynxlynx> yeah
[16:05:56] <pupnik> aah painkillers
[16:06:05] <pupnik> god bless big pharma
[16:35:39] --> danamin has joined #GemRB
[16:36:09] <-- danamin has left IRC (Client Quit)
[17:33:21] <fuzzie> hm, my music changes have caused rather mad consolespam when there's no music
[17:36:52] <pupnik> :)
[18:31:17] <fuzzie> so many bugs :(
[18:32:17] <CIA-2> gemrb: 03fuzzie * r6364 10/gemrb/trunk/gemrb/plugins/TISImporter/TISImp.cpp: don't assume a headerless TIS stream is invalid if it happens to begin with a T
[18:32:27] <fuzzie> that code is being kind of foolish in any case :| we shouldn't be playing guessing games on the stream
[18:32:37] <pupnik> mmm
[18:43:33] <fuzzie> TIS overlay colourkeying is annoying
[18:44:30] <fuzzie> i was hoping to fix Small Teeth Pass, but it turns out that there are tiles in the TIS with multiple shades of green, all of which should be colourkeyed. so that is non-trivial.
[18:44:55] <wjp> is that the green stuff I filed in the bug tracker some time ago?
[18:44:58] <fuzzie> yeah
[18:45:10] <fuzzie> i can fix the pool of water with a hack, it's all a single colour (4/226/4 i think)
[18:45:25] <fuzzie> but there are other tiles on the map with multiple shades of bright green
[18:45:46] <fuzzie> it is our oldest open bug :)
[18:46:36] <wjp> wow, 2006
[18:46:47] <wjp> "some time ago"...
[18:46:58] <fuzzie> I could add another hack and just rewrite all the tile pixels to a single colour, but I really don't know it's meant to work.
[18:50:28] <fuzzie> I added a comment to the bug, in any case.
[18:53:14] <fuzzie> And closed another one, hooray.
[18:53:46] <wjp> TNO death; yay :-)
[18:54:38] <fuzzie> i was pleasantly surprised to discover that someone already added the sarcastic party commentary upon death :)
[18:56:55] <fuzzie> death certainly seems to break mattinm's round code, though
[18:58:20] <fuzzie> It does InitRound for every tick on every previously-battling actor. I guess we should be force-stopping the rounds in circumstances like moving areas.
[19:04:29] <fuzzie> It really does look like pst rounds are 2 seconds.
[19:04:58] <fuzzie> Doesn't make much sense, you'd think someone would have noticed it before now! But all the discussions I can find on Google just assume that PST has 6 second rounds.
[20:26:53] <fuzzie> lynxlynxlynx: are you here?
[20:32:06] <fuzzie> nm, i confirmed myself
[20:36:40] <pupnik> nice
[20:39:28] <fuzzie> zefklop fixed the timers to use ROUND_SIZE
[20:39:35] <fuzzie> then changed ROUND_SIZE, breaking the timers
[20:42:07] <fuzzie> he said "I wrote it thinking that ROUND_SIZE could/should be changed :)" but i think almost all the code he wrote with ROUND_SIZE is wrong now :/
[20:48:36] --> danamin has joined #GemRB
[20:49:17] <danamin> Hi again, is there a call to set the character sound?
[20:59:45] <fuzzie> um
[21:01:31] <-- Gekz has left IRC (Remote closed the connection)
[21:01:32] <wjp> there's SetPlayerSound
[21:01:37] --> Gekz has joined #GemRB
[21:02:42] <fuzzie> that seems to be the right function, too
[21:17:23] <fuzzie> ok, i guess i'm changing all the ROUND_SIZE stuff to AI_UPDATE_TIME :/
[21:24:44] --> |Cable| has joined #gemrb
[21:25:16] <-- barra_library has left IRC ("Verlassend")
[21:38:57] <CIA-2> gemrb: 03fuzzie * r6365 10/gemrb/trunk/gemrb/plugins/ (10 files in 4 dirs): game timing/timers are counted in AI update ticks, not rounds
[21:39:01] <fuzzie> that is tested against the original game.
[21:39:22] <fuzzie> and gosh that took too long.
[21:46:34] <fuzzie> so: pst rounds are 2 seconds, bg1/bg2/iwd rounds 6 seconds, iwd2 rounds 7 seconds
[21:47:04] <lynxlynxlynx> i guess all the saves from the interim are bad
[21:47:20] <fuzzie> only a few, i think.
[21:47:36] <fuzzie> anything with a timer set is going to take 6 times too long.
[21:47:46] <fuzzie> otherwise it should be okay.
[21:48:23] <fuzzie> http://forums.gibberlings3.net/index.php?showtopic=9017 seems to imply that days are a different length in iwd2, among other things
[21:50:00] <fuzzie> the timing stuff definitely needs more research, but ROUND_SIZE was clearly wrong, where I changed it. maybe there are other bugs.
[21:53:09] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[21:54:55] <fuzzie> maybe the day length code is still wonky, don't know
[21:56:04] <lynxlynxlynx> i think it'll be better now
[21:56:16] <lynxlynxlynx> we definitely made an improvement since the last release
[21:56:31] --> Gekz has joined #GemRB
[21:57:03] <fuzzie> the bg1 manual says a game day is 'just over 2 hours long in real time'..
[21:58:00] <fuzzie> but the x86 machine here has been taken for actual useful work now, so i can't test
[22:07:57] <lynxlynxlynx> 5 min => 1 h; 24 h <= 2h
[22:08:10] <lynxlynxlynx> so i agree
[22:10:09] <-- pupnik has left IRC (Read error: 110 (Connection timed out))
[22:12:28] <danamin> Is it correct that BG1 has only 4 thief skills? (PICK_POCKETS,OPEN_LOCKS,FIND_TRAPS,MOVE_SILENTLY) and not (HIDE_IN_SHADOWS,DETECT_ILLUSION and SET_TRAPS)
[22:13:11] <danamin> skills 2da has all 7 but the chargen gui has only 4 slots and no scrollbar?
[22:13:25] <fuzzie> hm
[22:13:47] <fuzzie> the manual says that the bg1 stealth mode is 'a blind of Move Silently and Hide in Shadows'
[22:14:54] <fuzzie> and describes the other skills as "Pick Pockets, Open Locks and Remove Traps"
[22:14:59] <fuzzie> so my answer is 'probably'
[22:16:47] <danamin> OK, I'm going to sleep, the BG1 chargen rework is almost done, it'll probably be done tomorrow
[22:16:57] <fuzzie> night! good to hear :)
[22:17:12] <-- danamin has left IRC ()
[22:18:15] <lynxlynxlynx> yes, bg1 had stealth instead of separate skills
[22:20:25] <-- lynxlynxlynx has left IRC (Remote closed the connection)
[22:22:47] --> Edheldil has joined #gemrb
[22:22:47] --- ChanServ gives channel operator status to Edheldil
[22:25:52] --> pupnik has joined #gemrb
[22:26:08] --> barra_library has joined #gemrb
[22:56:14] --> pupnik_ has joined #gemrb
[22:56:49] <-- pupnik has left IRC (Read error: 110 (Connection timed out))
[23:23:43] --> pupnik__ has joined #gemrb
[23:24:20] <Edheldil> night all
[23:24:24] <-- Edheldil has left IRC (Remote closed the connection)
[23:36:10] <-- pupnik_ has left IRC (Read error: 110 (Connection timed out))
[23:50:11] <fuzzie> AttackOneRound is using AI_UPDATE_TIME, i expect this is wrong
[23:51:40] <fuzzie> otherwise time usage now looks consistent